MTU and Spyder on same circuit give very different readings

[GH]

To troubleshoot a new TED installation I have two pairs of CTs on one pair of wires coming out of a double breaker.  One CT pair is 400A split collar routed to the MTU and the other CT pair is 200A split collar routed to a spyder that is attached to that MTU.  Spyder settings combine those two CTs (ports 1 and 2).

The MTU pair and the Spyder pair give very different energy readings, Spyder is circa 5x larger and they move roughly in parallel but it varies a bit.  MTU Skp is less than 5%.

Any known issues that come to mind?  Is there a calibration setting required for the 400A CTs?

[pfletch101]

Things to check:
Are all the CTs properly closed round their respective wires?
Are the two MTU CTs oriented in the same way (red dots on both either towards or away from the breaker)?
Is there generation involved - can the direction of power flux through the breaker reverse under any reasonably likely circumstances? This will cause trouble because Spyder channels assume unidirectional flux.
Unless these checks bring up anything, you should probably raise a ticket on the main site. TED people are here quite frequently, but not necessarily every day. In my experience, tickets are responded to very promptly - usually within 24 hours. There have been some recent hardware changes requiring significant calibration corrections to be used for at least some combinations of CTs and ECCs.

[Support7]

You will have to give me exact readings to be able to judge the "5x" difference better but key differences in the MTU and Spyder that need to be noted when comparing them on the same circuit are:
MTU is rated at +/-2% while the Spyder is rated at +/-7% so that may explain some of the difference.
MTU samples readings thousands of times a second to give you the average of that second while the Spyder samples thousands of times a minute to give the average of the minute so that will explain some differences in readings as well.
Also the Spyder and the MTU handle the data differently when calculating the readings based on certain things like power factor and kva so realtime readings may not seem similar but the averages work out over time. Smaller loads will also affect the accuracy of readings between the 400A and 200A CT's so if your measuring less than 500w with the 200A CT with the Spyder than there is a drop off point.

Spyder Specs: http://www.theenergydetective.com/downloads/Spyder%20Spec%20rev%202.2.pdf
Spyder vs MTU: http://www.theenergydetective.com/downloads/spydercompare.pdf

Hope that helps but yeah for troubleshooting if it's outside of the +/-9% difference (MTU 2% and Spyder 7%) than you will need to submit a support ticket

Support Ticket: http://www.theenergydetective.com/ticket

[GH]

Thank you both for the replies.  I will recheck orientation and closure as you suggest.  There is generation involved, I am "sure" it comes in before the main panel so will recheck that as well.

Graph attached showing the circa 5x difference and generally parallel movement.

If the rechecks come up dry I will open a support ticket.

Thanks again.

[GH]

Posting the resolution, I swapped things out one by one and it turned out to be a bad MTU, bad in that it was returning inaccurate (way too high) spyder readings.

During the troubleshooting I learned that the latching split-collar and squeeze-type split-collar 200A CTs are calibrated differently, and MTU's with serial number 17xxxyy are set up to use the squeeze-type.  If you use a latching-type 200A CT instead the calibration factor is 2.61 (Ted Pro /Advanced/Modify MTU Calibration.

Thanks to David in support.